A scheme is described for locally Bayesian parameter updating in models structured as successions of component functions. The essential idea is to back-propagate the target data to interior modules, such that an interior component’s target is the input to the next component that maximizes the probability of the next component’s target. Each layer then does locally Bayesian learning. The approach assumes online trial-by-trial learning. The resulting parameter updating is not globally Bayesian but can better capture human behavior. The approach is implemented for an associative learning model that first maps inputs to attentionally filtered inputs and then maps attentionally filtered inputs to outputs. The Bayesian updating allows the associative model to exhibit retrospective revaluation effects such as backward blocking and unovershadowing, which have been challenging for associative learning models. The back-propagation of target values to attention allows the model to show trial-order effects, including highlighting and differences in magnitude of forward and backward blocking, which have been challenging for Bayesian learning models.

Abstract. We give families of examples where sharp rates of convergence to stationarity of the widely used Gibbs sampler are available. The examples involve standard exponential families and their conjugate priors. In each case, the transition operator is explicitly diagonalizable with classical orthogonal polynomials as eigenfunctions. Key words and phrases: Gibbs sampler, running time analyses, exponential families, conjugate priors, location families, orthogonal polynomials, singular value decomposition. 1.

Abstract. The use of simulation for high-dimensional intractable computations has revolutionized applied mathematics. Designing, improving and understanding the new tools leads to (and leans on) fascinating mathematics, from representation theory through micro-local analysis. 1.

A difficulty in the development and testing of psychological models is that they are typically evaluated solely on their ability to fit experimental data, with little consideration given to their ability to fit other possible data patterns. By examining how well model A fits data generated by model B, and vice versa (a technique that we call landscaping), much safer inferences can be made about the meaning of a models fit to data. We demonstrate the landscaping technique using four models of retention and 77 historical data sets, and show how the method can be used to (1) evaluate the distinguishability of models, (2) evaluate the informativeness of data in distinguishing between models, and (3) suggest new ways to distinguish between models. The generality of the method is demonstrated in two other research areas (information integration and categorization), and its relationship to the important notion of model complexity is discussed.

As the number of alerts generated by collaborative applications grows, users receive more unwanted alerts. FeedMe is a general alert management system based on XML feed protocols such as RSS and ATOM. In addition to traditional rule-based alert filtering, FeedMe uses techniques from machine-learning to infer alert preferences based on user feedback. In this paper, we present and evaluate a new collaborative naïve Bayes filtering algorithm. Using FeedMe, we collected alert ratings from 33 users over 29 days. We used the data to design and verify the accuracy of the filtering algorithm and provide insights into alert prediction. Categories and Subject Descriptors H.5.3 [Group and Organization Interfaces]: Collaborative

Does psi exist? In a recent article, Dr. Bem conducted nine studies with over a thousand participants in an attempt to demonstrate that future events retroactively affect people’s responses. Here we discuss several limitations of Bem’s experiments on psi; in particular, we show that the data analysis was partly exploratory, and that one-sided p-values may overstate the statistical evidence against the null hypothesis. We reanalyze Bem’s data using a default Bayesian t-test and show that the evidence for psi is weak to nonexistent. We argue that in order to convince a skeptical audience of a controversial claim, one needs to conduct strictly confirmatory studies and analyze the results with statistical tests that are conservative rather than liberal. We conclude that Bem’s p-values do not indicate evidence in favor of precognition; instead, they indicate that experimental psychologists need to change the way they conduct their experiments and analyze their data. Keywords: Confirmatory Experiments, Bayesian Hypothesis Test, ESP. In a recent article for Journal of Personality and Social Psychology, Bem (in press) presented nine experiments that test for the presence of psi. 1 Specifically, the experiments were designed to assess the hypothesis that future events affect people’s thinking and people’s behavior in the past (henceforth precognition). As indicated by Bem, precognition—if it exists—is an anomalous phenomenon, because it conflicts with what we know to be true about the word (e.g., weather forecasting agencies do not employ clairvoyants, casino’s 1 The preprint that this article is based on was downloaded September 25th, 2010, from

Abstract not found

EM-Algorithm and Gibbs sampler are two useful Bayesian simulation methods for parameter estimation of finite normal mixture model. The EM-Algorithm is an iterative estimate of maximum likelihood for incomplete data problem. Gibbs sampler is an approach of generating random sample from a multivariate distribution. We introduce and derive Dempster EM-Algorithm for the two-component normal mixture models to get the iterative computation estimates, also use data augmentation and general Gibbs sampler to get the sample from posterior distribution under conjugate prior. The estimate results from both simulation methods under two-component normal mixture model with unknown mean parameters are compared and the connections and differences between both methods are represented. Data set from astronomy is used for comparison. Acknowledgement I would like to thank my supervisor Silvelyn Zwanzig for the patience, guidance and encouragement that she always gave to me, not only in the thesis, but also in the whole procedure of my statistics studying. I would also like to thank my friend Han Jun for the the assistances of LATEX, thank Alena for the data source, and thank my parents for the spiritual and substantial support and wholesouled love they gave me all my life. At last I would like to thank the department of mathematics of Uppsala University for giving me the opportunity to study. Contents 1

Abstract—The field of complex biomechanical modeling has begun to rely on Monte Carlo techniques to investigate the effects of parameter variability and measurement uncertainty on model outputs, search for optimal parameter combinations, and define model limitations. However, advanced stochastic methods to perform data-driven explorations, such as Markov chain Monte Carlo (MCMC), become necessary as the number of model parameters increases. Here, we demonstrate the feasibility and, what to our knowledge is, the first use of an MCMC approach to improve the fitness of realistically large biomechanical models. We used a Metropolis–Hastings algorithm to search increasingly complex parameter landscapes (3, 8, 24, and 36 dimensions) to uncover underlying distributions of anatomical parameters of a “truth model” of the human thumb on the basis of simulated kinematic data (thumbnail location, orientation, and linear and angular velocities)

We examine financial constraints and forms of finance used for investment, by analysing survey data on 157 large privatised companies in Hungary and Poland for the period 1998 – 2000. The Bayesian analysis using Gibbs sampling is carried out to obtain inferences about the sample companies ’ access to finance from a model for categorical outcome. By applying alternative measures of financial constraints we find that foreign companies, companies that are part of domestic industrial groups and enterprises with concentrated ownership are all less constrained in their access to finance. Moreover, we identify alternative modes of finance since different corporate control and past performance characteristics influence the sample firms ’ choice of finance source. In particular, while being industry-specific, the access to domestic credit is positively associated with company size and past profitability. Industrial group members tend to favour bond issues as well as sells-offs of assets as appropriate types of finance for their investment programmes. Preferences for raising finance in the form of equity are associated with share concentration in a non-monotonic way, being most prevalent in those companies where the dominant owner holds 25%-49 % of shares. Close links with a leading bank not only increase the